基于小波分析的大坝观测数据异常值检测  

Outliers Detecting of Dam Observation Data Based on Wavelet Analysis

作　　者：徐洪钟[1] 吴中如[2] 李雪红[2] 施斌[1] 

机构地区：[1]南京大学地球科学系,江苏南京210093 [2]河海大学水利水电工程学院,江苏南京210098

出　　处：《水电能源科学》2002年第4期20-21,81,共3页Water Resources and Power

基　　金：国家自然科学基金资助项目(50139030)。

摘　　要：阐述了小波分析中小波变换、多分辨分析等基本原理;利用多分辨分析(小波多层分解)方法,对大坝观测数据的异常值进行了检测;通过工程实例验证了该方法的有效性。所采用的小波分析方法适合检测单个和多个异常值。Outliers detecting of dam observation data is a basic work for dam observation data analysis. Dam observation data is a time series made up of different frequency, and outliers of dam observation data can be detected based on wavelet analysis. Firstly, basic theories such as wavelet transform and multiresolution analysis are presented in this paper. Outliers detecting of dam observation data is performed using multiresolution analysis. Finally, this method is proved to be effective through an example. This method is suitable for single and multiple outliers. 

关 键 词：大坝观测数据 异常值 检测 小波分析 

分 类 号：TV698.1[水利工程—水利水电工程]